Systems Are the Message

A scoreboard that displays kills and damage tells the lobby what mattered tonight, whatever the mode’s objective claims. A ranked queue with an entry cost says losses are supposed to hurt here. A crafting system built on randomized drops says something too. On my portfolio page for Hextech Crafting I compressed it to one line: “agency of choice is given up for a reduction in cost.” I still think that sentence was the real design, and everything else we built was just the UI for it.

Designers like to believe the rhetoric lives in the writing (the lore, the tooltips, the dev blog), but it doesn’t. It lives in the incentives, the visibility decisions, and the prices. You’re making rhetorical moves with every system you ship whether you intend them or not, so you might as well make them on purpose.

And you can’t hide the message, because players audit. They datamine drop rates, spreadsheet ladder points, scrape match histories into third-party sites that outgrow your own client. A live game can’t keep a secret. All it gets to choose is whether the eventual reveal confirms what it said out loud or contradicts it. When the visible number and the real number disagree, players don’t file it as a tuning problem, they file it as a lie. I don’t mean any of that as cynicism about players. That kind of auditing is the medium’s native literacy, and it’s earned my respect.

New Champion Select taught me this at scale. The screen we shipped in January 2016 made two statements at once: pick intent said your plan matters, while the queue attached to it said what a ladder measures is negotiable. Players parsed the two separately, with total precision. They kept the first and burned the second down. Chapters 03 and 07 take that screen apart.

Run the tape on the last decade. Riot spent the spring of 2016 insisting solo queue wasn’t coming back, conceded a version of it by August, and replaced Dynamic Queue entirely for 2017. Respawn announced tap-strafing’s removal in 2021, walked it back under community fire, and ran a compressed rerun of the same loop in 2024. Apex’s Season 17 ranked rework put a decorative progress bar over a hidden rating and inflated Masters until the public mea culpa and the rebuild. The Season 22 battle pass restructure in 2024 drew on the order of thirty thousand negative Steam reviews and was fully reversed in sixteen days. Riot removed earnable Hextech chests in early 2025 and reinstated them inside two months.

That’s five cases across two studios in nine years. I watched some of them from inside the building and some from the blast radius next door, and the pattern held every time. These weren’t communication failures. The community managers did their jobs. Players revolted because they understood the change perfectly. The system said your time is worth less than your retention, or your rank is whatever we need it to be this season, and players heard it on day one and refused to accept it. Reversal-and-apology has become the industry’s de facto constitutional process, the only mechanism players have for vetoing a statement they never agreed to. Chapter 07 takes that process apart properly. For now it stands as the evidence for this preface’s claim: players hear what systems say, even (and especially) when the studio doesn’t know it’s speaking.

Honor 2.0 is the same hearing, just running in the other direction. It shipped in patch 7.13, July 2017: an end-of-game vote, one honor to give per game, on purpose. As proof that players respond to what a system says rather than what the marketing says, it’s the cleanest result I’ve ever been near. What drifted afterward, and what Riot reversed in its 2024 postmortem, gets an honest accounting in chapter 06.

Swink called the sensation of controlling a digital object the most overlooked phenomenon in HCI, and the field mostly stopped reading at a hundred milliseconds. The sensation of being governed by a digital system (measured by it, priced by it, recognized by it over months) is the same phenomenon at longer exposure, and it gets even less credit. A hitstop and a ranked split are both the game telling you what your actions meant. So Section 01’s chapters are ordered by timescale, shortest to longest, because the structure is the argument:

01.01 Game feel. The first 33 milliseconds. The smallest statements: hitstop, recoil, the weight of a death. A time-to-kill number is a philosophy of dying. I shipped inside two studios that held opposite ones, both coherent.

01.02 Game UX. The session. Can they do it, do they understand it, does it feel good? In that order, no skipping. Web UX serves a user who wants to leave. We serve a player who wants to stay, and that changes most of what the web taught us.

01.03 Onboarding. The first hours. Marketing is a promise the game has to keep, launch day is part of the tutorial, and somewhere in the first session there’s a moment where the player nods. Everything before that nod is running on credit.

01.04 Competitive systems. The season. A ladder is a contract about what gets measured and what being measured costs. Keeping that contract explicit is the whole job. That, and the one thing a matchmaker must never be allowed to optimize.

01.05 Metagame. The account. One match record, opened up, powers tier lists, mastery, social resumes, and a third-party ecosystem no first-party roadmap could buy. Also: what the battle pass genuinely fixed, and what it traded the dice for.

01.06 Social systems. The community. The scoreboard is a values statement, probably the purest expression of this preface. What a game makes visible, players optimize, so choose what you display like it’s doctrine (because it is).

01.07 Live ops. The years. Who owns a shipped system: you, or the people living in it? The patch is the product, a season is a promise structure, and a reversal budget is insurance you buy hoping never to learn why you needed it.

After that comes Section 02, Design Vision: the template I use to take a game from a blank page to a first playable, including the Inside-Out Model for deciding what to build first, next, and later. Then Section 03, Getting Unstuck: twenty-one process tricks for getting unstuck, in three suits. No theory, just moves.